anymodality

/

llava-v1.5-13b

like 4

Hello,
I have followed all the instructions deploy_llava.ipynb notebook and it all went without issues or errors. Until I reached the part where the inference happens:

data = {
    "image" : 'https://raw.githubusercontent.com/haotian-liu/LLaVA/main/images/llava_logo.png', 
    "question" : "Describe the image and color details.",
    # "max_new_tokens" : 1024,
    # "temperature" : 0.2,
    # "stop_str" : "###"
}

# request
output = predictor.predict(data)
print(output)

I got the following error:

ModelError: An error occurred (ModelError) when calling the InvokeEndpoint operation: Received client error (400) from primary with message "{
  "code": 400,
  "type": "InternalServerException",
  "message": "GET was unable to find an engine to execute this computation"
}

I tried using the endpoint directly, but it returned the same error.
Any idea on how I can overcome this?